A method and system for distinguishing cable insulation degradation types

Abstract

The present invention discloses a method and system for distinguishing the type of cable insulation degradation. The method includes: obtaining the current waveform in the power-on experiment; intercepting a picture of a synthetic wave containing at least one complete cycle in the current waveform; obtaining the harmonic characteristic value of the synthetic wave and the contribution of the harmonics in the synthetic wave; obtaining the corresponding degradation type according to the contribution of the harmonics in the synthetic wave; constructing a cable insulation degradation type discrimination model; associating and binding the picture of the synthetic wave containing at least one complete cycle with its corresponding degradation type as a training picture, training the cable insulation degradation type discrimination model, and obtaining a trained cable insulation degradation type discrimination model. The present application uses a picture segment of the synthetic wave as input, which can blur the influence of harmonics caused by non-degradation related reasons, and at the same time avoids a large amount of calculation, which can greatly improve the test efficiency and accuracy, and is advanced.

Description

Technical Field

[0001] The invention belongs to the technical field of cable fault prevention and diagnosis, and in particular relates to a method and system for distinguishing cable insulation degradation types. Background Art

[0002] With the rapid development of science and technology, more and more detection work needs to be assisted by computers to improve detection efficiency and quality. For example, in the work of cable fault prevention and fault diagnosis, after the cable is energized, the harmonic characteristic values ​​at the cable detection point are extracted, and then the harmonic characteristic values ​​are input into the database for automatic comparison operation. The database gives the contribution corresponding to the 2nd, 3rd, 4th, 5th...nth harmonics, and then the staff can look up the table to get the cable degradation type;

[0003] Although the above method can ultimately make a relatively accurate judgment on the degradation type of the cable, the process requires a large amount of calculation, and harmonics may also be generated in the cable current due to non-degradation related reasons. These harmonics may be extracted with characteristic values ​​and included in the database's judgment data on the contribution rate, which may affect the final judgment of the cable degradation type.

[0004] The concept of deep learning originates from the study of artificial neural networks. A multilayer perceptron with multiple hidden layers is a deep learning structure. Deep learning combines low-level features to form more abstract high-level representations of attribute categories or features to discover distributed feature representations of data. The motivation for studying deep learning is to establish a neural network that simulates the human brain for analytical learning. It imitates the mechanism of the human brain to interpret data, such as images, sounds, and text. The calculations involved in generating an output from an input can be represented by a flow graph: a flow graph is a graph that can represent calculations, in which each node represents a basic calculation and a calculated value, and the result of the calculation is applied to the value of the child nodes of this node. Consider such a set of calculations that can be allowed in each node and possible graph structure, and define a family of functions. Summary of the invention

[0005] The main purpose of the present invention is to overcome the shortcomings and deficiencies of the prior art and to provide a method and system for distinguishing the type of cable insulation degradation.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions:

[0007] One aspect of the present invention provides a method for distinguishing the type of cable insulation degradation, comprising the following steps:

[0008] Obtain the current waveform in the power-on experiment;

[0009] capturing a picture of a synthetic wave containing at least one complete cycle of the current waveform;

[0010] Obtaining harmonic characteristic values ​​of the synthesized wave and contribution of harmonics in the synthesized wave;

[0011] Obtaining the corresponding degradation type according to the contribution of the harmonics in the synthetic wave;

[0012] Construct a model to identify the type of cable insulation degradation;

[0013] Associating and binding the picture containing at least one complete cycle of the synthetic wave with its corresponding degradation type as a training picture, training the cable insulation degradation type discrimination model, and obtaining a trained cable insulation degradation type discrimination model;

[0014] The synthetic wave image of the power-on experiment of the cable to be identified is input into the trained cable insulation degradation type identification model to obtain the degradation type of the cable to be identified.

[0015] As a preferred technical solution, the size and format of the pictures containing at least one complete cycle of the synthetic wave used as training pictures remain consistent, and are all captured from a position just past the zero point of the synthetic wave.

[0016] As an optimal technical solution, the harmonic characteristic value of the synthesized wave is obtained by performing Fourier transform on the synthesized wave to obtain the fundamental wave and each harmonic of the synthesized wave; the contribution of the harmonics in the synthesized wave is the proportion of each harmonic in all decomposed waveforms of the synthesized wave.

[0017] As a preferred technical solution, the construction of the cable insulation degradation type discrimination model is specifically as follows:

[0018] The cable insulation degradation type discrimination model includes an input layer, several hidden layers and an output layer connected in sequence, as shown in the following formula:

[0019]

[0020] in, Represents the output space; represents the input space; a(·) represents the bending operation on the input space; W represents the weight matrix of each layer of the neural network; It indicates the operations of increasing dimension, decreasing dimension, enlarging, reducing and rotating the input space; +b indicates the operations of translating the input space.

[0021] As a preferred technical solution, the training of the cable insulation degradation type discrimination model is specifically as follows:

[0022] Acquire the synthetic wave image containing at least one complete cycle and associate and bind it with the corresponding degradation type as a training image;

[0023] Input the training images into the input layer of the cable insulation degradation type discrimination model, and adjust the parameters of the weight matrix W according to the output results of the output layer to minimize the loss function;

[0024] Repeat the above steps for training.

[0025] As a preferred technical solution, after adding disturbance to the training image, the image is added to the training set to train the cable insulation degradation type discrimination model, wherein the adding disturbance includes adding burrs or pixels to the training image.

[0026] Another aspect of the present invention provides a cable insulation degradation type discrimination system, which is applied to the above-mentioned cable insulation degradation type discrimination method, and includes a waveform acquisition module, a harmonic characteristic value and contribution calculation module, a model construction module and a model training module;

[0027] The waveform acquisition module is used to acquire the current waveform in the power-on experiment and to intercept a picture of a synthetic wave containing at least one complete cycle in the current waveform;

[0028] The harmonic characteristic value and contribution calculation module is used to obtain the harmonic characteristic value of the synthetic wave and the contribution of the harmonics in the synthetic wave;

[0029] The model building module is used to build a cable insulation degradation type discrimination model;

[0030] The model training module is used to obtain the corresponding degradation type according to the contribution of the harmonics in the synthetic wave; the image of the synthetic wave containing at least one complete cycle is associated and bound with the corresponding degradation type as a training image, and the cable insulation degradation type discrimination model is trained to obtain a trained cable insulation degradation type discrimination model;

[0031] The synthetic wave image of the power-on experiment of the cable to be identified is input into the trained cable insulation degradation type identification model to obtain the degradation type of the cable to be identified.

[0032] As an optimal technical solution, the harmonic eigenvalue and contribution calculation module is used to perform Fourier transform on the composite wave to obtain the fundamental wave and each harmonic of the composite wave, and then obtain the contribution of the harmonics in the composite wave; the contribution of the harmonics in the composite wave is the proportion of each harmonic in all decomposed waveforms of the composite wave.

[0033] As a preferred technical solution, the cable insulation degradation type discrimination model includes an input layer, several hidden layers and an output layer connected in sequence, as shown in the following formula:

[0034]

[0035] in, Represents the output space; represents the input space; a(·) represents the bending operation on the input space; W represents the weight matrix of each layer of the neural network; It indicates the operations of increasing dimension, decreasing dimension, enlarging, reducing and rotating the input space; +b indicates the operations of translating the input space.

[0036] As a preferred technical solution, the training of the cable insulation degradation type discrimination model is specifically as follows:

[0037] Acquire the synthetic wave image containing at least one complete cycle and associate and bind it with the corresponding degradation type as a training image;

[0038] Adding disturbance to the training image and then adding it to the training set, wherein adding disturbance includes adding burrs or pixels to the training image;

[0039] Input the images in the training set into the input layer of the cable insulation degradation type discrimination model, and adjust the parameters of the weight matrix W according to the output results of the output layer to minimize the loss function;

[0040] Repeat the above steps for training.

[0041] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0042] (1) This application establishes a cable insulation degradation type discrimination model to analyze the occurrence and content of different types of harmonics in the distortion waves generated by the cable during the aging process, thereby determining the thermal aging type and aging degree of the cable. With the improvement of the database, the accuracy of cable degradation type judgment can be improved;

[0043] (2) Compared with the traditional manual table lookup comparison and analysis of cable degradation, which is easily affected by slight changes in waveforms, has large errors, low efficiency, and large amount of calculation, the intelligent analysis model designed in this application uses the image fragments of the synthetic wave as input, which can blur the influence of harmonics caused by non-degradation related reasons, exclude harmonics caused by non-degradation reasons from being included in the judgment data of the database, and improve the accuracy of system analysis. At the same time, it avoids a large amount of calculation, which can greatly improve the test efficiency and accuracy, and is advanced. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 It is a flow chart of a method for distinguishing the type of cable insulation degradation according to an embodiment of the present invention;

[0045] Figure 2It is a schematic diagram of the fundamental wave and each harmonic obtained after the synthetic wave is Fourier transformed in an embodiment of the present invention;

[0046] Figure 3 It is a structural block diagram of a cable insulation degradation type discrimination system according to an embodiment of the present invention. DETAILED DESCRIPTION

[0047] In order to enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without making creative work are within the scope of protection of the present application.

[0048] Example

[0049] like Figure 1 As shown, this embodiment provides a method for distinguishing the type of cable insulation degradation, comprising the following steps:

[0050] S1. Obtain the current waveform in the power-on experiment, specifically:

[0051] Take different cables to conduct power-on experiments, collect current at the cable detection points, and obtain the current waveform within a certain period of time;

[0052] S2, capturing a picture of a synthetic wave containing at least one complete cycle of the current waveform;

[0053] Furthermore, the size and format of each training image need to be kept consistent. If the pixels of the training image are 800*800, then the test image subsequently input into the cable insulation degradation type discrimination model needs to also keep the pixels 800*800.

[0054] Furthermore, the starting point of the synthetic wave is selected as close to the zero point as possible, and is kept consistent in the selection of subsequent training images.

[0055] S3, obtaining the harmonic characteristic value of the synthesized wave and the contribution of the harmonics in the synthesized wave;

[0056] By performing Fourier transform on the synthesized wave, the fundamental wave and harmonics of the synthesized wave are obtained. Figure 2 This is an example diagram of the 3rd harmonic, 5th harmonic and fundamental wave obtained after Fourier transform of the synthetic wave. It is not the current waveform obtained in this acquisition, but is only used to illustrate that the waveform can be decomposed;

[0057] The contribution of the harmonics in the synthesized wave is the proportion of each harmonic in all decomposed waveforms of the synthesized wave.

[0058] S4, obtaining the corresponding degradation type according to the contribution of the harmonics in the synthetic wave;

[0059] After acquiring the current waveform and obtaining a sample image, the harmonic characteristic value of the synthetic wave in the sample image is acquired, and then the acquisition result is sent to a database pre-stored with harmonic contribution corresponding to various degradation types for comparison to obtain the degradation type;

[0060] When classifying sample images, the contribution of a certain harmonic feature value is used for determination:

[0061] (a) The deteriorated part of the power cable is the insulation of the main body;

[0062] If the contribution rates of the 3rd and 5th harmonics are both about 41%, and the contribution rates of the 4th and 2nd harmonics are both about 6%, then based on experience, it can be determined that the degradation type is the initial degradation type;

[0063] If the contribution rate of the 2nd harmonic is about 55%, the contribution rate of the 4th harmonic is about 16%, the contribution rate of the 3rd harmonic is about 9%, and the contribution rate of the 5th harmonic is about 6%, then based on experience, it can be determined that the degradation type is environmental degradation (mechanical damage);

[0064] If the contribution rate of the 5th harmonic is about 59%, the contribution rate of the 3rd harmonic is about 20%, the contribution rate of the 4th harmonic is about 8%, and the contribution rate of the 2nd harmonic is about 6%, then based on experience, it can be determined that the degradation type is environmental degradation (electrical damage);

[0065] If the contribution rate of the 5th harmonic is about 52%, the contribution rate of the 3rd harmonic is about 28%, the contribution rate of the 4th harmonic is about 7%, and the contribution rate of the 2nd harmonic is about 6%, then based on experience, it can be determined that the degradation type is the long-term degradation type;

[0066] (b) The deteriorated part of the power cable is the conductor of the main body;

[0067] If the contribution rate of the third harmonic is about 25%, the contribution rate of the fifth harmonic is about 24%, the contribution rate of the second harmonic is about 23%, and the contribution rate of the fourth harmonic is about 18%, then based on experience, it can be determined that the deteriorated part of the power cable is the conductor of the main body;

[0068] (c) The deteriorated part of the power cable is the protective layer of the main body;

[0069] If the contribution rate of the 2nd harmonic is about 39%, the contribution rate of the 4th harmonic is about 29%, the contribution rate of the 3rd harmonic is about 10%, and the contribution rate of the 5th harmonic is about 7%, then based on experience, it can be determined that the deteriorated part of the power cable is the protective layer of the main body;

[0070] (d) The deteriorated part of the power cable is the cable joint at the connection part;

[0071] If the contribution rate of the 7th harmonic is about 53%, the contribution rate of the 10th harmonic is about 15%, the contribution rate of the 9th harmonic is about 11%, the contribution rate of the 8th harmonic is about 7%, and the contribution rate of the 6th harmonic is about 5%, then based on experience, it can be determined that the degradation type is heating;

[0072] If the contribution rate of the 8th harmonic is about 35%, the contribution rate of the 7th harmonic is about 29%, the contribution rate of the 9th harmonic is about 13%, the contribution rate of the 10th harmonic is about 11%, and the contribution rate of the 6th harmonic is about 7%, then based on experience, it can be determined that the degradation type is contamination;

[0073] If the contribution rate of the 9th harmonic is about 33%, the contribution rate of the 8th harmonic is about 25%, the contribution rate of the 7th harmonic is about 21%, the contribution rate of the 10th harmonic is about 8%, and the contribution rate of the 6th harmonic is about 5%, then based on experience, it can be determined that the degradation type is cracking;

[0074] If the contribution rate of the 10th harmonic is about 30%, the contribution rate of the 7th harmonic is about 23%, the contribution rate of the 8th harmonic is about 17%, the contribution rate of the 9th harmonic is about 15%, and the contribution rate of the 6th harmonic is about 6%, then based on experience, it can be determined that the degradation type is deformation;

[0075] S5. Construct a cable insulation degradation type discrimination model;

[0076] The cable insulation degradation type discrimination model in this embodiment is a trained deep neural network; the neural network is composed of neural units, which can be specifically understood as a neural network with an input layer, several hidden layers and an output layer connected in sequence, the first layer is the input layer, the last layer is the output layer, and the layers in between are all hidden layers. Among them, a neural network with many hidden layers is called a deep neural network (DNN).

[0077] The work of each layer in the neural network can be described by a mathematical expression as follows:

[0078]

[0079] in, Represents the output space; represents the input space; a(·) represents the bending operation on the input space; W represents the weight matrix of each layer of the neural network; +b represents the operation of increasing / decreasing the dimension, enlarging / reducing the dimension, and rotating the input space; +b represents the operation of translating the input space. From a physical perspective, the work of each layer in a neural network can be understood as completing the transformation from input space to output space (i.e., from the row space to the column space of a matrix) through five operations (increasing / decreasing the dimension, enlarging / reducing the dimension, rotating, translating, and bending) on ​​the input space (a set of input vectors).

[0080] The word "space" is used to describe it because the object being classified is not a single thing, but a class of things. Space refers to the collection of all individuals of this class of things, where W is the weight matrix of each layer of the neural network, and each value in the matrix represents the weight value of a neuron in that layer. The matrix W determines the spatial transformation from the input space to the output space described above, that is, the W of each layer of the neural network controls how to transform the space. The purpose of training a neural network is to eventually obtain the weight matrices of all layers of the trained neural network. Therefore, the training process of a neural network is essentially about learning how to control spatial transformations, or more specifically, learning the weight matrix.

[0081] S6, associating and binding the picture containing at least one complete cycle of the synthetic wave with its corresponding degradation type as a training picture, training the cable insulation degradation type discrimination, and obtaining a trained cable insulation degradation type discrimination model;

[0082] S6.1. Obtain the image containing at least one complete cycle of the synthetic wave and associate it with the corresponding degradation type to serve as a training image;

[0083] S6.2, input the training image into the input layer of the cable insulation degradation type discrimination model, and adjust the parameters of the weight matrix W according to the output result of the output layer so that the loss function is minimized;

[0084] S6.3. Repeat the above steps for training.

[0085] Among them, regarding the loss function, in the process of training the neural network, because we hope that the output of the neural network is as close as possible to the value we really want to predict, we can compare the predicted value of the current network with the target value we really want, and then update the weight matrix of each layer of the neural network according to the difference between the two (of course, there is usually an initialization process before the first update, that is, pre-configuring parameters for each layer in the neural network). For example, if the predicted value of the network is high, adjust the weight matrix to make it predict lower, and keep adjusting until the neural network can predict the target value we really want. Therefore, it is necessary to predefine "how to compare the difference between the predicted value and the target value", which is the loss function or objective function, which are important equations used to measure the difference between the predicted value and the target value. Among them, taking the loss function as an example, the higher the output value loss of the loss function, the greater the difference, so the training of the neural network becomes a process of minimizing this loss as much as possible.

[0086] In particular, disturbances may be added to the training images and then added to the training set to train the cable insulation degradation type discrimination model, thereby improving the contribution assessment accuracy; the added disturbances include adding glitches or pixel points to the training images.

[0087] S7. Input the synthetic wave image of the power-on experiment of the cable to be identified into the trained cable insulation degradation type identification model to obtain the degradation type of the cable to be identified.

[0088] In some periods of time, harmonics and noise that are not caused by cable degradation or other reasons may exist in the synthetic wave, which will affect the extraction of harmonic characteristic values. However, when these noises are displayed in the form of pictures, the trained cable insulation degradation type discrimination model proposed by the present invention will not be affected by the specific values ​​of the synthetic wave, because slight changes in the waveform can be overcome by the cable insulation degradation type discrimination model. Therefore, such a classification method can improve the accuracy of judging the degradation type.

[0089] like Figure 3 As shown, in another embodiment of the present application, a cable insulation degradation type discrimination system is provided, the system comprising a waveform acquisition module, a harmonic characteristic value and contribution calculation module, a model construction module and a model training module;

[0090] The waveform acquisition module is used to acquire the current waveform in the power-on experiment and to intercept a picture of a synthetic wave containing at least one complete cycle in the current waveform;

[0091] The harmonic characteristic value and contribution calculation module is used to obtain the harmonic characteristic value of the synthetic wave and the contribution of the harmonics in the synthetic wave;

[0092] The model building module is used to build a cable insulation degradation type discrimination model;

[0093] The model training module is used to obtain the corresponding degradation type according to the contribution of the harmonics in the synthetic wave; the image of the synthetic wave containing at least one complete cycle is associated and bound with the corresponding degradation type as a training image, and the cable insulation degradation type discrimination model is trained to obtain a trained cable insulation degradation type discrimination model;

[0094] The synthetic wave image of the power-on experiment of the cable to be identified is input into the trained cable insulation degradation type identification model to obtain the degradation type of the cable to be identified.

[0095] Furthermore, the size and format of the pictures containing at least one complete cycle of the synthetic wave used as training pictures remain consistent, and are all captured from a position just past the zero point of the synthetic wave.

[0096] Furthermore, the harmonic characteristic value and contribution calculation module is used to perform Fourier transform on the synthesized wave to obtain the fundamental wave and each harmonic of the synthesized wave, and then obtain the contribution of the harmonics in the synthesized wave; the contribution of the harmonics in the synthesized wave is the proportion of each harmonic in all decomposed waveforms of the synthesized wave.

[0097] Furthermore, the cable insulation degradation type discrimination model includes an input layer, several hidden layers and an output layer connected in sequence, as shown in the following formula:

[0098]

[0099] in, Represents the output space; represents the input space; a(·) represents the bending operation on the input space; W represents the weight matrix of each layer of the neural network; It indicates the operations of increasing dimension, decreasing dimension, enlarging, reducing and rotating the input space; +b indicates the operations of translating the input space.

[0100] Furthermore, the training of the cable insulation degradation type discrimination model is specifically as follows:

[0101] Acquire the synthetic wave image containing at least one complete cycle and associate and bind it with the corresponding degradation type as a training image;

[0102] Adding disturbance to the training image and then adding it to the training set, wherein adding disturbance includes adding burrs or pixels to the training image;

[0103] Input the images in the training set into the input layer of the cable insulation degradation type discrimination model, and adjust the parameters of the weight matrix W according to the output results of the output layer to minimize the loss function;

[0104] Repeat the above steps for training.

[0105] It should be noted here that the system provided in the above embodiment is only illustrated by the division of the above functional modules. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure is divided into different functional modules to complete all or part of the functions described above. The system is a method for distinguishing the type of cable insulation degradation applied to the above embodiment.

[0106] It should be understood that the various parts of the present application can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0107] The above embodiments are preferred implementation modes of the present invention, but the implementation modes of the present invention are not limited to the above embodiments. Any other changes, modifications, substitutions, combinations, and simplifications that do not deviate from the spirit and principles of the present invention should be equivalent replacement methods and are included in the protection scope of the present invention.

Claims

1. A method for distinguishing the type of cable insulation degradation, characterized in that: The steps include: Obtain the current waveform in the power-on experiment; capturing a picture of a synthetic wave containing at least one complete cycle of the current waveform; Obtaining harmonic characteristic values ​​of the synthesized wave and contribution of harmonics in the synthesized wave; Obtaining the corresponding degradation type according to the contribution of the harmonics in the synthetic wave; Construct a model to identify the type of cable insulation degradation; Associating and binding the picture containing at least one complete cycle of the synthetic wave with its corresponding degradation type as a training picture, training the cable insulation degradation type discrimination model, and obtaining a trained cable insulation degradation type discrimination model; The synthetic wave image of the power-on experiment of the cable to be identified is input into the trained cable insulation degradation type identification model to obtain the degradation type of the cable to be identified.

2. A method for distinguishing the type of cable insulation degradation according to claim 1, characterized in that: The size and format of the pictures containing at least one complete cycle of the synthetic wave used as training pictures remain consistent, and are all captured from the position just past the zero point of the synthetic wave.

3. A method for distinguishing the type of cable insulation degradation according to claim 1, characterized in that: The method of obtaining the harmonic characteristic value of the synthesized wave is as follows: performing Fourier transform on the synthesized wave to obtain the fundamental wave and each harmonic of the synthesized wave; the contribution of the harmonics in the synthesized wave is the proportion of each harmonic in all decomposed waveforms of the synthesized wave.

4. A method for distinguishing the type of cable insulation degradation according to claim 1, characterized in that: The construction of the cable insulation degradation type discrimination model is specifically as follows: The cable insulation degradation type discrimination model includes an input layer, several hidden layers and an output layer connected in sequence, as shown in the following formula: in, Represents the output space; represents the input space; a(·) represents the bending operation on the input space; W represents the weight matrix of each layer of the neural network; It indicates the operations of increasing dimension, decreasing dimension, enlarging, reducing and rotating the input space; +b indicates the operations of translating the input space.

5. A method for distinguishing the type of cable insulation degradation according to claim 1, characterized in that: The training of the cable insulation degradation type discrimination model is specifically as follows: Acquire the synthetic wave image containing at least one complete cycle and associate and bind it with the corresponding degradation type as a training image; Input the training images into the input layer of the cable insulation degradation type discrimination model, and adjust the parameters of the weight matrix W according to the output results of the output layer to minimize the loss function; Repeat the above steps for training.

6. A method for distinguishing the type of cable insulation degradation according to claim 1, characterized in that: After adding disturbance to the training image, the image is added to the training set to train the cable insulation degradation type discrimination model, wherein the adding disturbance includes adding burrs or pixels to the training image.

7. A cable insulation degradation type identification system, characterized in that: A cable insulation degradation type discrimination method applied to any one of claims 1 to 6, comprising a waveform acquisition module, a harmonic characteristic value and contribution calculation module, a model construction module and a model training module; The waveform acquisition module is used to acquire the current waveform in the power-on experiment and to intercept a picture of a synthetic wave containing at least one complete cycle in the current waveform; The harmonic characteristic value and contribution calculation module is used to obtain the harmonic characteristic value of the synthetic wave and the contribution of the harmonics in the synthetic wave; The model building module is used to build a cable insulation degradation type discrimination model; The model training module is used to obtain the corresponding degradation type according to the contribution of the harmonics in the synthetic wave; the image of the synthetic wave containing at least one complete cycle is associated and bound with the corresponding degradation type as a training image, and the cable insulation degradation type discrimination model is trained to obtain a trained cable insulation degradation type discrimination model; The synthetic wave image of the power-on experiment of the cable to be identified is input into the trained cable insulation degradation type identification model to obtain the degradation type of the cable to be identified.

8. A cable insulation degradation type identification system according to claim 7, characterized in that: The harmonic characteristic value and contribution calculation module is used to perform Fourier transform on the synthesized wave to obtain the fundamental wave and each harmonic of the synthesized wave, and then obtain the contribution of the harmonics in the synthesized wave; the contribution of the harmonics in the synthesized wave is the proportion of each harmonic in all decomposed waveforms of the synthesized wave.

9. A cable insulation degradation type identification system according to claim 7, characterized in that: The cable insulation degradation type discrimination model includes an input layer, several hidden layers and an output layer connected in sequence, as shown in the following formula: in, Represents the output space; represents the input space; a(·) represents the bending operation on the input space; W represents the weight matrix of each layer of the neural network; It indicates the operations of increasing dimension, decreasing dimension, enlarging, reducing and rotating the input space; +b indicates the operations of translating the input space.

10. A cable insulation degradation type identification system according to claim 7, characterized in that: The training of the cable insulation degradation type discrimination model is specifically as follows: Acquire the synthetic wave image containing at least one complete cycle and associate and bind it with the corresponding degradation type as a training image; Adding disturbance to the training image and then adding it to the training set, wherein adding disturbance includes adding burrs or pixels to the training image; Input the images in the training set into the input layer of the cable insulation degradation type discrimination model, and adjust the parameters of the weight matrix W according to the output results of the output layer to minimize the loss function; Repeat the above steps for training.

Images